Data Dashboard

The Data Dashboard is an add-on bundle that turns Crow into a lightweight data exploration platform. Browse database schemas, write SQL queries, visualize results with charts, and publish findings as blog posts.

Overview

Install the Data Dashboard bundle to get a new panel in the Crow's Nest with four tabs:

Tab	Purpose
Schema Explorer	Browse tables, columns, types, and relationships across connected databases
SQL Editor	Write and run queries with syntax highlighting and result tables
Charts	Build visualizations from query results (bar, line, pie, scatter)
Case Studies	Combine queries, charts, and narrative into publishable case studies

Getting Started

Install the bundle:

"Install the data dashboard"

Or via CLI:

crow bundle install data-dashboard
crow bundle start data-dashboard

The Data Dashboard panel appears in the Nest sidebar after installation.

Schema Explorer

The Schema Explorer shows every database registered as a data backend. Select a database from the dropdown to see its tables, columns, data types, and foreign key relationships.

Use it to understand unfamiliar datasets before writing queries. The explorer reads schema metadata only — it never touches your data.

SQL Editor

Write SQL queries against any registered database. Features:

• Syntax highlighting and basic autocompletion
• Result table with sortable columns and row counts
• Save queries with a name and description for reuse
• Export results as CSV or JSON

SELECT county, COUNT(*) as filings
FROM tax_returns
WHERE year = 2025
GROUP BY county
ORDER BY filings DESC
LIMIT 20;

Run the query with the Execute button or Ctrl+Enter.

Saved Queries

Saved queries persist in the Crow database. Access them from the SQL Editor's sidebar. Each saved query records:

• Name and description
• The SQL text
• Which database it targets
• When it was last run

Charts

Select a saved query or run an ad-hoc query, then switch to the Charts tab to visualize the results.

Supported chart types:

• Bar — Compare categories (e.g., filings by county)
• Line — Show trends over time (e.g., monthly submissions)
• Pie — Show proportions (e.g., credit type distribution)
• Scatter — Explore relationships between two numeric columns

Charts are rendered with Chart.js. Configure axis labels, colors, and titles in the chart editor. Save charts alongside their source queries.

Case Studies

A case study combines multiple queries, charts, and written analysis into a single document. Use case studies to tell a data story.

Creating a Case Study

1. Run your queries and build your charts
2. Open the Case Studies tab and click New Case Study
3. Add sections — each section can be narrative text (Markdown), a saved query with its result table, or a chart
4. Arrange sections by dragging them into order
5. Preview the rendered case study

Publishing to Blog

Case studies can be published directly to your Crow blog:

"Publish my tax analysis case study to the blog"

The AI converts the case study into a blog post, embedding charts as images and query results as formatted tables. The original case study remains editable — republish after updates.

Safety Model

The Data Dashboard enforces strict safety boundaries:

• Read-only by default — Only SELECT queries are allowed. INSERT, UPDATE, DELETE, and DDL statements are blocked unless you explicitly enable write mode for a specific database.
• Path restrictions — SQLite databases must be within allowed directories (~/.crow/data/, registered backend paths). No access to system databases or files outside the sandbox.
• Query timeouts — Queries are killed after 30 seconds to prevent runaway operations.
• No remote execution — Queries run locally against registered backends. Federation queries go through the gateway proxy with the same safety checks on the remote side.

Next Steps

• Data Backends — Register external databases
• Data Sharing — Share databases with other Crow users
• Data Dashboard Architecture — Technical deep dive
• Extending the Dashboard — Add chart types and exporters